Treatment of ulnar neuropathy at the elbow: cost-utility analysis

Abstract

Purpose: The choice of surgical treatment for ulnar neuropathy at the elbow (UNE) remains controversial. A cost-utility analysis was performed for 4 surgical UNE treatment options. We hypothesized that simple decompression would emerge as the most cost-effective strategy.

Methods: A cost-utility analysis was performed from the societal perspective. A decision analytic model was designed comparing 4 strategies: (1) simple decompression followed by a salvage surgery (anterior submuscular transposition) for a poor outcome, (2) anterior subcutaneous transposition followed by a salvage surgery for a poor outcome, (3) medial epicondylectomy followed by a salvage surgery for a poor outcome, and (4) anterior submuscular transposition. A poor outcome when anterior submuscular transposition was the initial surgery was considered an end point in the model. Preference values for temporary health states for UNE, the surgical procedures, and the complications were obtained through a time trade-off survey administered to family members and friends who accompanied patients to physician visits. Probabilities of clinical outcomes were derived from a Cochrane Collaboration meta-analysis and a systematic MEDLINE and EMBASE search of the literature. Medical care costs (in 2009 U.S. dollars) were derived from Medicare reimbursement rates. The model estimated quality-adjusted life-years and costs for a 3-year time horizon. A 3% annual discount rate was applied to costs and quality-adjusted life-years. Incremental cost-effectiveness ratios were calculated, and sensitivity analyses performed.

Results: Simple decompression as an initial procedure was the most cost-effective treatment strategy. A multi-way sensitivity analysis varying the preference values for the surgeries and a model structure sensitivity analysis varying the model assumptions did not change the conclusion. Under all evaluated scenarios, simple decompression yielded incremental cost-effectiveness ratios less than US$2,027 per quality-adjusted life-year.

Conclusions: Simple decompression as an initial treatment option is cost-effective for UNE according to commonly used cost-effectiveness thresholds.

Type of study/level of evidence: Economic and Decision Analysis III.

APPENDIX B:

Time trade-off survey design. A Four versions of the time trade-off surveys with 7 to 9 scenarios each were designed. Both the scenarios and sequences of the scenarios were varied to reduce anchoring bias (see “Materials and Methods” for details). B Respondents were randomized to 1 of the 4 surveys. A total of 117 respondents completed 1 of the 4 surveys. Fifteen respondents’ responses were excluded for the following reasons: (1) inability to understand the questions based on their comments, (2) responses were mostly missing, or (3) the ordering of the health state was illogical. A total of 102 respondents’ responses were considered valid and were used to calculate utilities.

FIGURE 1:

Schematic of decision tree for reference case analysis. *After salvage surgery, possible complications following a poor outcome include no complication, superficial wound infection, scar tenderness beyond 4 months, and continued paresthesias with worsening UNE (ie, claw). †After simple decompression, anterior subcutaneous transposition, and anterior submuscular transposition, possible complications that might have caused or followed a poor outcome include no complication, scar tenderness beyond 4 months, superficial wound infection, persistent paresthesias in the ring and small fingers, and neuromas of the medial brachial or antebrachial cutaneous nerves. ‡After medial epicondylectomy, possible complications that might have caused or followed a poor outcome include no complication, scar tenderness beyond 4 months, superficial wound infection, persistent paresthesias in the ring and small fingers, neuromas of the medial brachial or antebrachial cutaneous nerves, flexor-pronator weakness, and valgus instability. §An additional branch for revision surgery following a neuroma complication was also modeled in the tree. ††After simple decompression, anterior subcutaneous transposition, anterior submuscular transposition, and salvage surgery, possible complications following a good outcome include no complication, scar tenderness beyond 4 months, and superficial wound infection. ‡‡Possible complications following a good outcome include no complication, scar tenderness beyond 4 months, superficial wound infection, flexor-pronator weakness, and valgus instability.

FIGURE 2:

A, B Sample time trade-off question.

FIGURE 3:

Incremental cost-effectiveness ratio equation.

FIGURE 4:

Schematic of decision tree for model structural sensitivity analysis. A revision surgery following a poor outcome for anterior submuscular transposition was modeled to account for the possibility that some surgeons perform a re-exploration of a poor outcome after an initial surgery (red dotted branch). *After salvage surgery, possible complications following a poor outcome include no complication, superficial wound infection, scar tenderness beyond 4 months, and continued paresthesias with worsening UNE (ie, claw). †After simple decompression, anterior subcutaneous transposition, and anterior submuscular transposition, possible complications following a poor outcome include no complication, scar tenderness beyond 4 months, superficial wound infection, persistent paresthesias in the ring and small fingers, and neuromas of the medial brachial or antebrachial cutaneous nerves. ‡After medial epicondylectomy, possible complications following a poor outcome include no complication, scar tenderness beyond 4 months, superficial wound infection, persistent paresthesias in the ring and small fingers, neuromas of the medial brachial or antebrachial cutaneous nerves, flexor-pronator weakness, and valgus instability. ††After simple decompression, anterior subcutaneous transposition, anterior submuscular transposition, and salvage surgery, possible complications following a good outcome include no complication, scar tenderness beyond 4 months, and superficial wound infection. ‡‡Possible complications following a good outcome include no complication, scar tenderness beyond 4 months, superficial wound infection, flexor-pronator weakness, and valgus instability.

FIGURE 5:

Cost-effectiveness plane. A The results of a cost-effectiveness analysis are typically presented by the 4-quadrant cost-effectiveness plane. A comparator procedure or intervention is indicated at the origin of the plane, and the alternative procedures are plotted relative to the comparator. Each quadrant is labeled as the NW, NE, SE, or SW quadrant. The southern quadrants are cost-saving, and interventions that fall into the SE quadrant are considered dominant; that is, it costs less and has greater total effectiveness. These interventions are typically accepted. However, interventions that fall into the SW quadrant are debated because there is a trade-off. Although it is cost saving, it is less effective. Interventions that fall into the NW quadrant are dominated; that is, they cost more and are less effective. These interventions are typically rejected. Finally, interventions that fall into the NE quadrant are also debated, given that there is a trade-off. Although they are more effective, they are also more costly. The dotted line represents the cost-effectiveness threshold value ($/QALY). Interventions that fall under the threshold in the NE quadrant are considered cost-effective. B The results for the reference case analysis are presented in the cost-effectiveness plane. The origin represents anterior submuscular transposition. Medial epicondylectomy was a dominated strategy (higher cost, lower effectiveness) and fell into the NW quadrant. It should thus be rejected. However, both simple decompression and anterior subcutaneous transposition fell into the NE quadrant, indicating that there is a trade-off (higher costs but also greater effectiveness). Abbreviations: NE, northeast; NW northwest; SE, southeast; SW, southwest.